1 / 34

Statistics for All: Nearer Our Destination or Slip Sliding Away?

Statistics for All: Nearer Our Destination or Slip Sliding Away?. Dick Scheaffer Ann Watkins. The Three Commandments. Emphasize statistical thinking. More data and concepts. Foster active learning. Defining Our Problems.

burke
Download Presentation

Statistics for All: Nearer Our Destination or Slip Sliding Away?

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Statistics for All: Nearer Our Destination or Slip Sliding Away? • Dick Scheaffer • Ann Watkins

  2. The Three Commandments • Emphasize statistical thinking • More data and concepts • Foster active learning

  3. Defining Our Problems • What particular goals will there be toward which the leading spirits of coming generations will strive?

  4. Problem #1 Decide what all citizens need to know about statistics.

  5. #1 What should every student understand by the end of high school? A. How to collect, organize, and display data B. Correlation and regression C. Variability and that it’s possible to generalize from a sample to a population, with some error D. Inference (conceptual only): margin of error, “statistical tie” E. Construct confidence intervals and perform significance tests

  6. #1 Audience Vote n = 162 (A vote for C, for example, would include A and B.) A. How to collect, organize, and display data 16% B. Correlation and regression 2% C. Variability and that it’s possible to generalize from a sample to a population, with some error 46% D. Inference (conceptual only): margin of error, “statistical tie” 30% E. Construct confidence intervals and perform significance tests 6%

  7. Problem #2 Statistics has not caught on in K-12 (AP Statistics excepted).

  8. EMPHASIS INDEX Robert Reys, University of Missouri Ratio of the emphasis placed on the topic by 39 middle school teachersto the emphasis placed on the topic by the text

  9. #2How should we resolve the problem that statistics has not caught on in K-12? A. Stay the course B. Coordinate our current grassroots efforts C. Change emphasis to showing teachers how to fit stat into their math curriculum D. Apply pressure from the top down, working with mathematicians E. Remove statistics from the math curriculum F. Give it up

  10. #2Audience Vote n = 163 A. Stay the course 1% B. Coordinate our current grassroots efforts 16% C. Change emphasis to showing teachers how to fit stat into their math curriculum 67% D. Apply pressure from the top down, working with mathematicians 6% E. Remove statistics from the math curriculum 10% F. Give it up 1%

  11. Problem #3 Much of the statistics in state frameworks and assessments is awful.

  12. California Framework Students know the central limit theorem and can use it to obtain approximations for probabilities in problems of finite sample spacesin which the probabilities are distributed binomially. (12th grade) Identify different ways of selecting a sample (e.g., convenience sampling, responses to a survey, random sampling) and which method makes a sample more representative for a population. (6th grade) Identify data that represent sampling errors and explain whythe sample (and the display) might be biased. (6th grade)

  13. According to the circle graph— A more Children’s books were borrowed than Romance and Science Fiction combined. B more than half of the books borrowed were Children’s, Mysteries, and Art combined. C more Mysteries were borrowed than Art and Science Fiction combined. D more than half of the books borrowed were Romance, Mysteries, and Science Fiction combined. California High School Exit Examination, 2003 Released Data Analysis Question Smithburg town library wanted to see what types of books were borrowed most often.

  14. #3 How should we improve the statistics in K-12 state documents? A. Stay the course—largely ignore state documents B. Coordinate our grassroots efforts C. Apply pressure from the top down D. Move statistics from the math curriculum E. Educate the people who write them

  15. #3 Audience Vote n = 142 A. Stay the course—largely ignore state documents 2% B. Coordinate our grassroots efforts 18% C. Apply pressure from the top down 13% D. Move statistics from the math curriculum 6% E. Educate the people who write them 61%

  16. Problem #4 Even with GAISE, we will not have a grade-by-grade sequence of objectives for K-12 statistics.

  17. #4 Who should take the leadership and responsibility for developing grade-by-grade standards? A. “Individual” efforts such as GAISE B. ASA C. NCTM D. CBMS or MSEB E. State departments of education F. It’s not important to develop them

  18. #4 Audience Vote n = 140 A. “Individual” efforts such as GAISE 4% B. ASA 50% C. NCTM 23% D. CBMS or MSEB 7% E. State departments of education 16% F. It’s not important to develop them <1%

  19. Problem #5 Some mathematicians say that there is no room for statistics in K-12 and young kids can’t understand it anyway. Statistics can wait.

  20. #5 Should we build a research base towards substantiating that we can and must teach statistics early and often? • A. Absolutely, such a research base does not exist • B. Such a research base partially exists—more work is needed • C. A research base does exist but it needs to be summarized and publicized • D. Such research is important but I don’t know what has been done • E. A research base is not important

  21. #5 Audience Vote n = 140 • A. Absolutely, such a research base does not exist 11% • B. Such a research base partially exists—more work is needed 24% • C. A research base does exist but it needs to be summarized and publicized 9% • D. Such research is important but I don’t know what has been done 56% • E. A research base is not important 1%

  22. Problem #6 We haven’t changed our expectations for students entering the intro college course.

  23. Remedial Statistics? The Implications of the Changing Secondary School Curriculum Statistics for the Twenty-First Century MAA, 1992

  24. #6 Should we work towards assuming prerequisite knowledge for intro stats? A. No B. Collect, summarize, and display data C. Correlation and regression D. Variability and that it’s possible to generalize from a sample to a population, with some error E. Inference (conceptual only): margin of error, “statistical tie” F. Construct confidence intervals and perform significance tests

  25. #6 Audience Vote n = 155 (A vote for C, for example, would include A and B.) A. No 15% B. Collect, summarize, and display data 34% C. Correlation and regression 11% D. Variability and that it’s possible to generalize from a sample to a population, with some error 31% E. Inference (conceptual only): margin of error, “statistical tie” 6% F. Construct confidence intervals and perform significance tests 3%

  26. Problem #7 The basic content of the introductory statistics course has not changed in 30 years.

  27. Changes in the Last 30 Years • Leading books now written by statisticians • More real data • Activity-based • Technology-using • More graphics and simulation • Regression diagnostics • Cleaned up theory

  28. #7 How should we teach inference in intro stats? A. Continue to teach the “standard” methods B. Move to computer-intensive methods: randomization tests, bootstrap C. This isn’t our decision—it depends on what client disciplines use in follow-up courses D. Don’t teach any method — teach concepts and intelligent use of software

  29. #7 Audience Vote n = 152 A. Continue to teach the “standard” methods 16% B. Move to computer-intensive methods: randomization tests, bootstrap 22% C. This isn’t our decision—it depends on what client disciplines use in follow-up courses 13% D. Don’t teach any method — teach concepts and intelligent use of software 49%

  30. Problem #8 We still aren’t teaching statistics to all that many college students.

  31. Enrollments in elementary statistics in math and stat departments and in TYC math programs ÷ college population (in thousands) Fall 1990: 171/13819 = .012 Fall 2000: 264/15312 = .017 In Two-Year Colleges only Fall 1980: 28/4526 = .006 Fall 2000: 74/5847 = .013

  32. #8 What is the main problem preventing more students from taking more statistics? A. Students think statistics isn’t important B. Students think statistics is too difficult C. Reform has backfired—students want less active learning and fewer concepts D. Other disciplines don’t value statistical thinking in their students E. Too many people teaching intro stat who shouldn’t be

  33. #8 Audience Vote n = 133 A. Students think statistics isn’t important 15% B. Students think statistics is too difficult 32% C. Reform has backfired—students want less active learning and fewer concepts 3% D. Other disciplines don’t value statistical thinking in their students 29% E. Too many people teaching intro stat who shouldn’t be 21%

More Related